2021
DOI: 10.1007/s10489-021-02323-4
|View full text |Cite
|
Sign up to set email alerts
|

Using known nonself samples to improve negative selection algorithm

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

0
6
0

Year Published

2023
2023
2024
2024

Publication Types

Select...
6

Relationship

0
6

Authors

Journals

citations
Cited by 6 publications
(6 citation statements)
references
References 30 publications
0
6
0
Order By: Relevance
“…The compared algorithms include the real-valued negative selection algorithm (RVNSA) [28], improved negative selection algorithm (INSA) [29], known nonself (KN) [30], and adaptive immunoregulation negative selection algorithm (AI-NSA) [31]. It is assumed that 40% of the nonself set is known and used for INSA and KN to optimize the detector distribution.…”
Section: A Performance Comparison Of Up-to-date Nsas On a Benchmark D...mentioning
confidence: 99%
See 2 more Smart Citations
“…The compared algorithms include the real-valued negative selection algorithm (RVNSA) [28], improved negative selection algorithm (INSA) [29], known nonself (KN) [30], and adaptive immunoregulation negative selection algorithm (AI-NSA) [31]. It is assumed that 40% of the nonself set is known and used for INSA and KN to optimize the detector distribution.…”
Section: A Performance Comparison Of Up-to-date Nsas On a Benchmark D...mentioning
confidence: 99%
“…In NSAs, generating a large number of detectors is typically time-consuming because each generated detector must undergo self-tolerance and be checked against the existing detectors to remove redundancies. To verify the performance of the proposed algorithm, the same initial self-set is used with other algorithms, including RVNSA [28], INSA [29], KN [30], and AINSA [31]. The average increment of each round of candidate detectors is 100, and the maximum number of detectors is 58292, which is the total number of detectors generated by the proposed algorithm.…”
Section: ) Detector Generationmentioning
confidence: 99%
See 1 more Smart Citation
“…To reduce the time complexity of the detector calculation, Yang et al [30] applied the antigen spatial density to calculate the low-dimensional subspace of densely aggregated antigens which generates detectors directly in these subspaces. Fouladvand et al [31] improved the real-valued negative selection algorithm based on Delaunay triangular dissection (dnyNSA) to generate detectors with more rational locations and sizes. Li et al [32] employed the known nonself as the candidate detector center to generate the detector and thus efectively improve the detection rate.…”
Section: Detector Matching Tolerancementioning
confidence: 99%
“…Li and Chen [33] used the Monte Carlo method to calculate the overlap volume of the hypersphere and proposed a nonself-covering calculation method based on confdence estimation. Fouladvand et al [31] compared the randomly generated pattern with the self-space GMM and retained the low probability random pattern as a detector. Yang et al [30] applied "antibody inhibition rate" instead of "expected coverage" as the termination condition.…”
Section: Hole Repairmentioning
confidence: 99%